In my previous two pieces in this series, I talked a bit about how startups are an exercise in controlled chaos, and how building an enduring franchise means getting very comfortable with ambiguity. That’s easy to say, and easy to nod your head along to. But it’s much, much harder to truly understand the full extent of unexpected situations that can pop up — and how to handle them.

As Mike Tyson said, “Everyone has a plan until they get punched in the mouth.” And startups are all about getting punched in the mouth. Except in startup land, you rarely even know who or what is throwing the punch, which direction it’s coming from, or even that you’ve stepped into the ring.

Continuing the thread from the previous pieces, I want to share some of the punches we were dealt building NerdWallet, and some of the lessons I’ve picked up over the years that I apply when working with founders within the BTV portfolio. We built The Mint around these lessons, helping fintech startups navigate the zero-to-one phase and arm themselves for their own journeys. 

Two steps forward, one step back 

On Memorial Day weekend in 2010, my wife and I were leaving New York to pursue my startup dreams in San Francisco. We’d packed all our stuff into a moving truck, then we, my NerdWallet co-founder, Tim, and a bunch of our other friends took a trip to Barbados as a goodbye, and perhaps, our last vacation for a while.

Right when we got there, Lifehacker released a feature on us – a positive review on the depth of cards and banks featured in our credit card reviews. What a way to start our vacation! It was an exciting moment, the first real validation of what we spent so many tireless months putting together. We’d actually been picked up by Money Magazine and by the New York Times at this point, but Lifehacker was our people, and it sent more traffic to us on the first day than both of those combined.

It was a big moment for our young company– until our website crashed. 

The cheap dedicated server we rented from some random company we found online, and maintained ourselves, buckled under the weight of our newfound traffic. You see, at this point, we were basically just some PHP spaghetti code that we FTP’ed to a computer in a room somewhere, matched with a SQL database that we manually updated from an Excel spreadsheet. None of it was configured for speed or scale. And we didn’t have an engineering team, or DevOps, or anything like that. It was just us.

As Tim says, “Apparently, client-side rendering and database indexing didn’t matter when you had two visitors a day."

So, Tim and I sat in our swim trunks, on the patio, with our laptops out, trying to figure out how to get our server back online from the Caribbean Sea. Some way to start our vacation. Our early win had turned into a disaster.